On Artificial Intelligence: An interview with Tugba Akinci D'Antonoli - AI Blog - ESR | European Society of Radiology %

In our latest interview, we spoke to Tugba Akinci D’Antonoli, a radiology resident at Cantonal Hospital Baselland and a researcher at the University of Basel, Switzerland. D’Antonoli is a member of the 2023–2025 trainee editorial board for Radiology: Artificial Intelligence, a scientific editorial board member for European Radiology and Diagnostic and Interventional Radiology and is […]

Performance of dual-energy subtraction in contrast-enhanced mammography for three different manufacturers: a phantom study - European Radiology Experimental

Background Dual-energy subtraction (DES) imaging is critical in contrast-enhanced mammography (CEM), as the recombination of low-energy (LE) and high-energy (HE) images produces contrast enhancement while reducing anatomical noise. The study's purpose was to compare the performance of the DES algorithm among three different CEM systems using a commercial phantom. Methods A CIRS Model 022 phantom, designed for CEM, was acquired using all available automatic exposure modes (AECs) with three CEM systems from three different manufacturers (CEM1, CEM2, and CEM3). Three studies were acquired for each system/AEC mode to measure both radiation dose and image quality metrics, including estimation of measurement error. The mean glandular dose (MGD) calculated over the three acquisitions was used as the dosimetry index, while contrast-to-noise ratio (CNR) was obtained from LE and HE images and DES images and used as an image quality metric. Results On average, the CNR of LE images of CEM1 was 2.3 times higher than that of CEM2 and 2.7 times higher than that of CEM3. For HE images, the CNR of CEM1 was 2.7 and 3.5 times higher than that of CEM2 and CEM3, respectively. The CNR remained predominantly higher for CEM1 even when measured from DES images, followed by CEM2 and then CEM3. CEM1 delivered the lowest MGD (2.34 ± 0.03 mGy), followed by CEM3 (2.53 ± 0.02 mGy) in default AEC mode, and CEM2 (3.50 ± 0.05 mGy). The doses of CEM2 and CEM3 increased by 49.6% and 8.0% compared with CEM1, respectively. Conclusion One system outperformed others in DES algorithms, providing higher CNR at lower doses. Relevance statement This phantom study highlighted the variability in performance among the DES algorithms used by different CEM systems, showing that these differences can be translated in terms of variations in contrast enhancement and radiation dose. Key Points DES images, obtained by recombining LE and HE images, have a major role in CEM. Differences in radiation dose among CEM systems were between 8.0% and 49.6%. One DES algorithm achieved superior technical performance, providing higher CNR values at a lower radiation dose. Graphical Abstract

AI-assisted double reading system able to identify missed findings on chest radiographs following repot authorization - AI Blog - ESR | European Society of Radiology %

Our study evaluated an AI-assisted double reading system for chest radiographs in two different hospital settings. The system analysed both the radiograph and the corresponding radiologist’s report to detect potential missed findings. Among 25,104 chest radiographs, clinically relevant missed findings were confirmed in 0.1% of cases, primarily consisting of unreported lung nodules, pneumothoraces, and consolidations. […]

Positive MRI and ISUP GG1 on initial prostate biopsy? Reassessing baseline MRI is key

Misdiagnosis in breast imaging: a statement paper from European Society Breast Imaging (EUSOBI)—Part 1: The role of common errors in radiology in missed breast cancer and implications of misdiagnosis - European Radiology

Importance Misdiagnosis in breast imaging can have significant implications for patients, healthcare providers, and the healthcare system as a whole. Observations Some of the potential implications of misdiagnosis in breast imaging include delayed diagnosis or false reassurance, which can result in a delay in treatment and potentially a worse prognosis. Misdiagnosis can also lead to unnecessary procedures, which can cause physical discomfort, anxiety, and emotional distress for patients, as well as increased healthcare costs. All these events can erode patient trust in the healthcare system and in individual healthcare providers. This can have negative implications for patient compliance with screening and treatment recommendations, as well as overall health outcomes. Moreover, misdiagnosis can also result in legal consequences for healthcare providers, including medical malpractice lawsuits and disciplinary action by licensing boards. Conclusion and relevance To minimize the risk of misdiagnosis in breast imaging, it is important for healthcare providers to use appropriate imaging techniques and interpret images accurately and consistently. This requires ongoing training and education for radiologists and other healthcare providers, as well as collaboration and communication among healthcare providers to ensure that patients receive appropriate and timely care. If a misdiagnosis does occur, it is important for healthcare providers to communicate with patients and provide appropriate follow-up care to minimize the potential implications of the misdiagnosis. This may include repeat imaging, additional biopsies or other procedures, and referral to specialists for further evaluation and management. Key Points Question What factors most contribute to and what implications stem from misdiagnosis in breast imaging? Findings Ongoing training and education for radiologists and other healthcare providers, as well as interdisciplinary collaboration and communication, is paramount. Clinical relevance Misdiagnosis in breast imaging can have significant implications for patients, healthcare providers, and the entire healthcare system.

Validation of a multi-parameter algorithm for personalized contrast injection protocol in liver CT - European Radiology Experimental

Background In liver computed tomography (CT), tailoring the contrast injection to the patient’s specific characteristics is relevant for optimal imaging and patient safety. We evaluated a novel algorithm engineered for personalized contrast injection to achieve reproducible liver enhancement centered on 50 HU. Methods From September 2020 to August 31, 2022, CT data from consecutive adult patients were prospectively collected at our multicenter premises. Inclusion criteria consisted of an abdominal CT referral for cancer staging or follow-up. For all examinations, a web interface incorporating data from the radiology information system (patient details and examination information) and radiographer-inputted data (patient fat-free mass, imaging center, kVp, contrast agent details, and imaging phase) were used. Calculated contrast volume and injection rate were manually entered into the CT console controlling the injector. Iopamidol 370 mgI/mL or Iohexol 350 mgI/mL were used, and kVp varied (80, 100, or 120) based on patient habitus. Results We enrolled 384 patients (mean age 61.2 years, range 21.1–94.5). The amount of administered iodine dose (gI) was not significantly different across contrast agents (p = 0.700), while a significant increase in iodine dose was observed with increasing kVp (p < 0.001) and in males versus females (p < 0.001), as expected. Despite the differences in administered iodine load, image quality was reproducible across patients with 72.1% of the examinations falling within the desirable range of 40–60 HU. Conclusion This study validated a novel algorithm for personalized contrast injection in adult abdominal CT, achieving consistent liver enhancement centered at 50 HU. Relevance statement In healthcare’s ongoing shift towards personalized medicine, the algorithm offers excellent potential to improve diagnostic accuracy and patient management, particularly for the detection and follow-up of liver malignancies. Key Points The algorithm achieves reproducible liver enhancement, promising improved diagnostic accuracy and patient management in diverse clinical settings. The real-world study demonstrates this algorithm’s adaptability to different variables ensuring high-quality liver imaging. A personalized algorithm optimizes liver CT, improving the visibility, conspicuity, and follow-up of liver lesions. Graphical Abstract

MRI findings of newly present benign focal hepatic observations following chemotherapy: distinct features in early- and late-term follow-up - European Radiology

Objective To evaluate gadoxetic acid-enhanced (Gd-EOB-DTPA) MRI features of newly detected benign focal hepatic observations after chemotherapy. Methods In this retrospective single-center case-control study, we enrolled a cohort of 43 cancer patients with 93 newly detected benign focal hepatic observations after chemotherapy between January 2010 and December 2020. We evaluated several parameters including the delay of occurrence after chemotherapy, imaging features, and imaging follow-up. These parameters were compared with those observed in a control group comprising 34 patients with 93 hepatic metastases. Results For focal hepatic observations occurring at early-term follow-up (delay of occurrence after chemotherapy, median 3 months, range 1–6 months) with 22 patients encompassing 45 lesions, most lesions exhibited an ill-defined margin on HBP images (64.4%), negative on diffusion-weighted images (84.4%), mottled hypo-intensity on hepatobiliary phase images (88.9%), and undistorted vessels traversing the lesions (80.0%). Follow-up imaging indicated that 91.9% of these lesions resolved within 4–20 months. For focal hepatic observations occurring at late-term follow-up (delay of occurrence after chemotherapy, median 34 months, range 12–60 months) with 21 patients encompassing 48 lesions, which were diagnosed as focal nodular hyperplasia (FNH)-like lesions based on MRI features. A hepatobiliary ring enhancement was observed in 56.3% of lesions, and 66.7% of patients showed an increase in lesion size and/or number during follow-up imaging. Conclusion Focal hepatic observations occurring at early-term and late-term follow-ups after chemotherapy have distinctive imaging features at Gd-EOB-DTPA-MRI. Early-term focal observations tend to resolve spontaneously, whereas FNH-like lesions can increase in size and number during follow-up. Key Points Question Focal benign liver lesions related to chemotherapy-induced hepatic injury were reported in recent years, often leading to confusion with metastasis and resulting in misdiagnosis. Findings Chemotherapy-induced focal hepatic observations identified during early- and late-term follow-up exhibit distinct imaging characteristics on Gd-EOB-DTPA-MRI and demonstrate varying temporal changes. Clinical relevance Chemotherapy-induced hepatic observations can be differentiated from metastasis based on Gd-EOB-DTPA MRI findings and their temporal changes. A deeper understanding of their findings can avoid unnecessary biopsies or surgical resections.

Deep learning-based defacing tool for CT angiography: CTA-DEFACE - European Radiology Experimental

Abstract The growing use of artificial neural network (ANN) tools for computed tomography angiography (CTA) data analysis underscores the necessity for elevated data protection measures. We aimed to establish an automated defacing pipeline for CTA data. In this retrospective study, CTA data from multi-institutional cohorts were utilized to annotate facemasks (n = 100) and train an ANN model, subsequently tested on an external institution’s dataset (n = 50) and compared to a publicly available defacing algorithm. Face detection (MTCNN) and verification (FaceNet) networks were applied to measure the similarity between the original and defaced CTA images. Dice similarity coefficient (DSC), face detection probability, and face similarity measures were calculated to evaluate model performance. The CTA-DEFACE model effectively segmented soft face tissue in CTA data achieving a DSC of 0.94 ± 0.02 (mean ± standard deviation) on the test set. Our model was benchmarked against a publicly available defacing algorithm. After applying face detection and verification networks, our model showed substantially reduced face detection probability (p < 0.001) and similarity to the original CTA image (p < 0.001). The CTA-DEFACE model enabled robust and precise defacing of CTA data. The trained network is publicly accessible at www.github.com/neuroAI-HD/CTA-DEFACE . Relevance statement The ANN model CTA-DEFACE, developed for automatic defacing of CT angiography images, achieves significantly lower face detection probabilities and greater dissimilarity from the original images compared to a publicly available model. The algorithm has been externally validated and is publicly accessible. Key Points The developed ANN model (CTA-DEFACE) automatically generates facemasks for CT angiography images. CTA-DEFACE offers superior deidentification capabilities compared to a publicly available model. By means of graphics processing unit optimization, our model ensures rapid processing of medical images. Our model underwent external validation, underscoring its reliability for real-world application. Graphical Abstract

Unlocking the future of leukodystrophy diagnosis: the promise and challenges of quantitative MRI